Peta-scale high-perfomlance computing systems are increasingly built with heterogeneous CPU and GPU nodes to achieve higher power efficiency and computation throughput. While providing unprecedented capabilities to co...Peta-scale high-perfomlance computing systems are increasingly built with heterogeneous CPU and GPU nodes to achieve higher power efficiency and computation throughput. While providing unprecedented capabilities to conduct computational experiments of historic significance, these systems are presently difficult to program. The users, who are domain experts rather than computer experts, prefer to use programming models closer to their domains （e.g., physics and biology） rather than MPI and OpenME This has led the development of domain-specific programming that provides domain-specific programming interfaces but abstracts away some performance-critical architecture details. Based on experience in designing large-scale computing systems, a hybrid programming framework for scientific computing on heterogeneous architectures is proposed in this work. Its design philosophy is to provide a collaborative mechanism for domain experts and computer experts so that both domain-specific knowledge and performance-critical architecture details can be adequately exploited. Two real-world scientific applications have been evaluated on TH-IA, a peta-scale CPU-GPU heterogeneous system that is currently the 5th fastest supercomputer in the world. The experimental results show that the proposed framework is well suited for developing large-scale scientific computing applications on peta-scale heterogeneous CPU/GPU systems.展开更多
Within the model, a definition of novel saturation and dilution degree is presented. A novel expression of equilibrium constant is derived by including interactions between components to reversible processes running b...Within the model, a definition of novel saturation and dilution degree is presented. A novel expression of equilibrium constant is derived by including interactions between components to reversible processes running between the components particles under a dynamic equilibrium. For partition equilibria, three types of basic processes are proposed and corresponding partition isotherms were derived. The isotherms are applied to the components partition between two phases for main heterogenous systems types. For construction and prediction ternary phase diagrams, a new method, the Component Binding by Saturation Model （CBSM） method, is proposed. Applying the partition isotherms derived, new extraction, evaporation and adsorption isotherms are expressed. Directly, or after an approximation, from the new adsorption isotherms, well-known adsorption isotherms are obtained.展开更多
基金Project(61170049) supported by the National Natural Science Foundation of ChinaProject(2012AA010903) supported by the National High Technology Research and Development Program of China
文摘Peta-scale high-perfomlance computing systems are increasingly built with heterogeneous CPU and GPU nodes to achieve higher power efficiency and computation throughput. While providing unprecedented capabilities to conduct computational experiments of historic significance, these systems are presently difficult to program. The users, who are domain experts rather than computer experts, prefer to use programming models closer to their domains （e.g., physics and biology） rather than MPI and OpenME This has led the development of domain-specific programming that provides domain-specific programming interfaces but abstracts away some performance-critical architecture details. Based on experience in designing large-scale computing systems, a hybrid programming framework for scientific computing on heterogeneous architectures is proposed in this work. Its design philosophy is to provide a collaborative mechanism for domain experts and computer experts so that both domain-specific knowledge and performance-critical architecture details can be adequately exploited. Two real-world scientific applications have been evaluated on TH-IA, a peta-scale CPU-GPU heterogeneous system that is currently the 5th fastest supercomputer in the world. The experimental results show that the proposed framework is well suited for developing large-scale scientific computing applications on peta-scale heterogeneous CPU/GPU systems.
文摘Within the model, a definition of novel saturation and dilution degree is presented. A novel expression of equilibrium constant is derived by including interactions between components to reversible processes running between the components particles under a dynamic equilibrium. For partition equilibria, three types of basic processes are proposed and corresponding partition isotherms were derived. The isotherms are applied to the components partition between two phases for main heterogenous systems types. For construction and prediction ternary phase diagrams, a new method, the Component Binding by Saturation Model （CBSM） method, is proposed. Applying the partition isotherms derived, new extraction, evaporation and adsorption isotherms are expressed. Directly, or after an approximation, from the new adsorption isotherms, well-known adsorption isotherms are obtained.
